Single nucleotide polymorphisms in ZNRD1-AS1 increase cancer risk in an Asian population

Single nucleotide polymorphisms (SNPs) in human zinc ribbon domain containing 1 antisense RNA 1 (ZNRD1-AS1) have been associated with cancer development. In this meta-analysis, we more precisely estimated the associations between three expression quantitative trait loci SNPs in ZNRD1-AS1 (rs3757328, rs6940552, and rs9261204) and cancer susceptibility. The data for three SNPs were extracted from eligible studies, which included 5,293 patients and 5,440 controls. Overall, no significant associations between SNPs in ZNRD1-AS1 (rs3757328, rs6940552, and rs9261204) and cancer risk were observed. However, in further subgroup analyses based on cancer type, we found that the A allele of rs3757328 increased the risk of some cancer in both allele contrast (OR = 1.15, 95% CI = 1.05 – 1.25) and recessive models (OR = 1.79; 95% CI = 1.33 – 2.41). The A allele of rs6940552 and the G allele of rs9261204 also increased the risk of some cancer in an Asian population in allele contrast (OR = 1.17, 95% CI = 1.08 – 1.26, and OR = 1.25, 95% CI = 1.16 – 1.34, respectively) and recessive models (OR = 1.44, 95% CI = 1.18 – 1.77, and OR = 1.49; 95% CI = 1.23 – 1.80, respectively). Thus, rs3757328, rs6940552, and rs9261204 in ZNRD1-AS1 are all associated with increased some cancer risk in an Asian population.


INTRODUCTION
Long noncoding RNAs (lncRNAs) are a class of RNAs greater than 200 nucleotides in length that are not translated into proteins [1].The expression of lncRNAs is cell type-and tissue-dependent, which distinguishes them from protein-coding genes [2].The secondary structures of the lncRNAs can dictate their functions in various cellular processes and diseases [3].Some lncRNAs activate the oncogenic signaling pathways to drive cancer phenotypes [4].For example, lncRNA HULC promotes the epithelialto-mesenchymal transition phenotype and tumorigenesis in both pancreatic and gastric cancer cells [5,6].
Single nucleotide polymorphisms (SNPs) in lncRNAs can also promote cancer development and progression.For example, the TT genotype of rs12826786 in HOTAIR was found to increase breast cancer susceptibility [7].Expression quantitative trait loci (eQTLs) in the lncRNA CARD8 are susceptibility markers for cervical cancer [8].The C/T genotype of rs3787016 in the lncRNA POLR2E was associated with a decreased risk of esophageal squamous cell carcinoma [9].Collectively, these data indicate SNPs in lncRNAs have important roles in tumorigenesis and as prognostic biomarkers.
Human zinc ribbon domain containing 1 (ZNRD1) is involved in the development of multiple cancers [10].Interestingly, three SNPs in the lncRNA ZNRD1-AS1 (rs3757328, rs6940552, and rs9261204), which lies in the upstream region of the ZNRD1 gene, were found to inhibit ZNRD1 expression and decrease the risk of cervical cancer [11].However, several studies have demonstrated that eQTLs in ZNRD1-AS1 increased the risk of hepatocellular carcinoma (HCC) [12,13] and lung cancer [10].These conflicting results for rs3757328, rs6940552, and rs9261204 in ZNRD1-AS1 need to be further studied.Therefore, we investigated the effects of

Study characteristics
Four case-control articles [10][11][12][13], which included 5,293 cases and 5,440 controls, were included in our metaanalysis.All of the eligible studies were comprised of Asian populations.In one study, Li et al, investigated both lung cancer and bladder cancer.We therefore considered this study as two independent studies in our analysis (Table 1).The studies were all published between June 15, 2015 and June 30, 2016.The sample size range was 1,000 to 3,067.Finally, the studies investigated distinct tumor types (HCC, lung cancer, bladder cancer, and cervical cancer, Table 1).
All studies explored the relationships between SNPs in ZNRD1-AS1 (rs3757328, rs6940552, and rs9261204) and cancer risk.The genotyping methods included TaqMan in one study, PCR-restriction fragment length polymorphism in two studies, and the Sequenom MassARRAY iPLEX platform in two studies (Table 1).The methodological quality of each study was evaluated using the Newcastle-Ottawa Scale (NOS).All the studies scored at least 6 on this scale (Supplementary Table 1).
The genotypes for rs3757328, rs6940552, and rs9261204 included AA, GA, and GG.Hardy-Weinberg equilibrium test statistics indicated that the probability of the null hypothesis for most of the genotypes was correct (Table 2).

Publication bias
Potential publication biases was evaluated using the Egger's and Begg's tests.The results provided statistical evidence for the absence of publication bias in all models (Table 3).

DISCUSSION
SNPs in lncRNAs contribute to the development of various cancers [7,8,14].In this study, we analyzed the associations between three SNPs (rs3757328, rs6940552, and rs9261204) in ZNRD1-AS1 and cancer risk.Our data indicated that these SNPs are correlated with an increased risk of several cancers in an Asian population.
The ZNRD1 protein contains two zinc ribbon domains [15].It catalyzes the transcription of DNA into RNA and is a potential therapeutic target for various diseases [16,17].Reduced ZNRD1 expression was observed in human gastric cancer [18,19].Interestingly, ZNRD1 was found to suppress CDK4, Cyclin D1, and p21    recessive and B. allele contrast.www.impactjournals.com/oncotarget[20] and inhibit the growth of gastric cancer and leukemia cells in vitro [17,21].Previous studies have indicated that ZNRD1-AS1 contribute to tumorigenesis through negative regulation of the ZNRD1 gene [10].EQTLs analysis has demonstrated that SNPs in ZNRD1-AS1 regulate of ZNRD1 expression [22,23].Three SNPs in ZNRD1-AS1 (rs3757328, rs6940552 and rs9261204) have been associated with an increased risk of several cancers.Wen et al. [12] reported three of the SNPs (rs3757328, rs7769930, and rs694055) were associated with an increased risk of HCC.Li et al. [10] demonstrated that the G allele of rs9261204 increased the risk of lung cancer by 1.45-fold compared to the A allele.Nevertheless, these three SNPs (rs3757328, rs7769930, and rs694055) in ZNRD1-AS1 decreased the risk of cervical cancer [11].In this meta-analysis, we evaluated the effect of these three SNPs on the risk of cancer among the subgroups by cancer type.We found that the A allele of SNP rs3757328, A allele of SNP rs6940552, and G allele of SNP rs9261204 in ZNRD1-AS1 were associated with increased risk of some cancer types (HCC, lung cancer, and bladder cancer) except cervical cancer.
The meta-analysis had limitations.First, only five individual studies (focused on HCC, lung cancer, bladder cancer, and cervical cancer) were included in our analysis, which impacted the quality of our results.Second, our analysis was limited to individuals of Asian descent.Therefore, the effects of the SNPs on non-Asian populations are not yet clear, and further studies are necessary to confirm our results.
In conclusion, our data indicate that three SNPs in ZNRD1-AS1 were correlated with an increased risk of several cancers.These results must be further evaluated in large-scale, randomized controlled trials involving different ethnic populations and cancers.

Search strategy
We searched the PubMed, Embase, and Web of Science databases for studies performed prior to June 30, 2016 that reported an association between SNPs in ZNRD1-AS1 and cancer risk.This comprehensive literature search was performed using free-text words combined with Medical Subject Headings, such as "ZNRD1", "ZNRD1-AS1", or "lncRNA" and "cancer", "carcinoma", "tumor", "tumour", or "neoplasm" and "polymorphism", "variation", "variant", "SNP", "mutation", or "genotype".The references cited in the retrieved articles were also reviewed to identify additional eligible studies (Supplementary Figure 1).

Inclusion and exclusion criteria
The study inclusion criteria were the following: (1) case-control design; (2)

Data extraction
Two investigators (PYW and JHL) independently extracted the data from each study, including the surname of the first author, publication year, type of cancer, numbers of cases and controls, ethnicity, genotype platform, and SNP genotype.Disagreement was resolved through a discussion with a third reviewer (YML).

Quality assessment
The methodological quality of each eligible study was evaluated using the NOS.Each study was evaluated based on the selection, comparability, and exposure scores.Summary scores ranging from 0 to 9 points were calculated.Higher score were indicative of a lower risk of bias.

Statistical analysis
Allele contrast, dominant, recessive, homozygous, and heterozygous models were used to analyze the associations between SNPs in ZNRD1-AS1 and cancer risk.We calculated ORs and 95% CIs in order to estimate the strength of the associations.The significance of the ORs was determined using Z tests.Heterogeneity between studies was assessed using the Chi square-based Q statistic.A random effects (DerSimonian-Laird method) or fixed effect (Mantel-Haenszel method) model was used to calculate pooled effect estimates in the presence (P < 0.10) or absence (P > 0.10) of heterogeneity, and subgroup analysis by cancer type was further performed.Sensitivity analysis was performed by excluding one study at a time and recalculating the risk effect.Begg's and Egger's tests were performed to evaluate publication bias.Data analysis was performed using the Stata software, version 12.0 (Stata Corporation; College Station, TX, USA).A P value < 0.05 was considered statistically significant.

Figure 2 :
Figure 2: Forest plot of cancer risk associated with ZNRD1-AS1 polymorphism rs6940552.Models represented in A. recessive and B. allele contrast.

Figure 3 :
Figure 3: Forest plot of cancer risk associated with ZNRD1-AS1 polymorphism rs9261204.Models represented in A. recessive and B. allele contrast.

Figure 1 :
Figure 1: Forest plot of cancer risk associated with ZNRD1-AS1 polymorphism rs3757328.Models represented in A.

Table 1 : Characteristics of studies on the association between SNPs in ZNRD1-AS1 and cancer Author Year Ethnicity Cases Controls Type of cancer Single Nucleotide Polymorphisms
Abbreviations: HCC, hepatocellular carcinoma; PCR-RFLP, polymerase chain reaction-restriction fragment length polymorphism.www.impactjournals.com/oncotargetAS1wassignificantly associated with an increased risk of some cancer types (HCC, lung cancer, and bladder cancer) except cervical cancer (occurs only in women) both in the recessive model (Figure1A, Recessive model: OR = 1.79; 95% CI = 1.33 -2.41, P = 0.569 for the heterogeneity test, I 2 = 0.0%) and additive genetic model (Figure